Marketplace agencies usually notice the capacity problem too late. It does not arrive as a dramatic operational crisis. It arrives as a Slack message at 17:42: “Can we quickly check why Walmart sales are down?” Then another client asks why Amazon spend stopped at noon. Then a third client wants a new TikTok Shop launch plan by Friday. Everyone is busy, the team is talented, and the agency still somehow loses the week to exports, screenshots and status calls.
The named mistake I see most often is planning agency capacity by client count. A founder says, “One senior strategist can handle six clients, one specialist can handle ten ad accounts, and reporting takes half a day.” That looks sensible in a planning sheet. It fails in marketplace work because effort is not driven by client count. Effort is driven by SKU complexity, marketplace mix, catalog quality, stock volatility, retail media spend, and how often the client needs a commercial decision.
My stance: marketplace agency capacity planning should not start with hours. It should start with decision load. If your software cannot show which clients create urgent profit decisions this week, it is not a capacity system. It is a prettier timesheet.
This guide is for marketplace agencies in Germany, the US and similar mature ecommerce markets with five or more employees. At that stage, growth starts to hurt in a very specific way: the agency can sell more retainers than the team can service well. The answer is not always hiring faster. Often the better answer is building a capacity model around profit signals, automation and client operating cadence before another account manager inherits a portfolio held together by heroic spreadsheet energy.
What current agency software advice gets right
The research landscape is useful. General agency software guides from platforms like Kantata, Teamwork, Ravetree and Birdview correctly focus on resource planning, utilization, scopes, timelines, rate cards and delivery margin. That matters. Agencies do need to know whether a strategist is booked at 110%, whether the retainer covers the hours used, and whether project work is quietly eating profit.
Marketplace-specific vendors add another important layer. MerchantSpring’s agency pages talk about scaling the agency without rebuilding reporting before every client meeting, with advertising, profit and operational context in one governed foundation. ChannelEngine explains marketplace management software through product content, pricing, inventory, orders, reporting and analytics. Productsup’s agency content is strong on scalable feed operations, onboarding pressure and AI-ready product discovery. Pacvue connects retail media execution with sales, inventory and profitability signals. Rithum emphasizes listing management, inventory buffers, order flows and listing-level profitability. KwickMetrics and SellerSonar both describe the agency pain of disconnected Amazon and Walmart workflows, multi-client reporting, SKU-level profit and operational complexity.
That is all directionally right. The best competitors understand that marketplace agencies do not only need task management. They need connected commerce data.
But most advice still misses the agency operator’s hardest question: which client work deserves senior attention this week? Not which task is overdue. Not which dashboard looks red. Not which client is loudest. Which decisions, if delayed, will cost the client money and consume agency margin?
The capacity problem is not “too many clients”
A normal digital agency can often approximate capacity with channels and deliverables. Ten SEO retainers, five paid search accounts, three website projects. Marketplace agencies do not get that luxury because two clients with the same retainer can create completely different operating loads.
Imagine Northstar Marketplace Studio in Berlin. The agency has 11 people and 18 retained clients. On paper, the portfolio looks manageable: six Amazon-only clients, seven Amazon plus Kaufland clients, three bol.com expansion projects and two Walmart accounts for US brands. Average retainer: €4,800 per month. The founder expects each account manager to own six clients.
Now look at decision load. Client A has 280 SKUs, stable stock, 12 active ad campaigns and a clean catalog. Client B has 4,600 SKUs, 38% of listings missing attributes on one marketplace, shared inventory across Amazon.de and Kaufland, and €18,000 monthly retail media spend. Both count as “one client”. They are not one unit of capacity.
If Client B’s hero SKU has €14 contribution margin, a 22% break-even ACOS and only nine days of stock, the account manager needs a decision today: reduce bids, shift budget to a substitute SKU, pause a promotion, or ask the client to replenish. If that decision waits until Friday reporting, the client may lose €1,200 in avoidable ad spend or sell through stock that would have protected a better margin channel. That is capacity expressed in money, not hours.
Plan capacity around decision load, not hours
A useful marketplace agency capacity model has four layers.
1. Portfolio complexity
Count the things that create operational load: marketplaces, countries, SKUs, parent-child structures, fulfillment methods, feed rules, retail media networks, currencies and client stakeholders. A 500-SKU Amazon account with FBA and one decision maker is not the same as a 500-SKU account split across Amazon Vendor, Walmart Marketplace, Shopify and Mirakl retailers.
2. Commercial risk
Prioritize the work that can damage profit quickly: low-margin hero SKUs, campaigns above break-even ACOS, stockouts, return-rate spikes, suppressed listings, Buy Box loss, price changes and marketplace fees. This is where generic project management tools struggle. They can tell you a task is due. They cannot tell you that a €900 ad budget should be paused because the SKU has no margin permission left.
3. Client cadence
Some clients need a monthly board pack. Others need weekly trading calls, daily budget checks during promotions and fast decisions from finance. Capacity planning should include meeting load, approval speed and reporting expectations. A slow-approving client consumes capacity differently from a commercially mature client that gives clear guardrails upfront.
4. Automation coverage
Finally, measure what software has genuinely removed from the team. Automated report scheduling is helpful. Automated alerts are better. Automated bid and budget recommendations connected to contribution margin, stock cover and channel role are better again. The point is not to automate the relationship. It is to stop using senior people for repeatable data assembly.
A simple capacity score agencies can actually use
Here is a practical starting model. Give every client a weekly capacity score from 1 to 5 in five categories:
- Marketplace spread: 1 for one marketplace, 5 for four or more marketplaces with different fulfillment or currency logic.
- SKU and catalog complexity: 1 for under 100 stable SKUs, 5 for 3,000+ SKUs or frequent feed corrections.
- Ad and retail media risk: 1 for low spend or conservative campaigns, 5 for aggressive spend across Amazon, Walmart, bol, TikTok Shop or retailer media.
- Profit volatility: 1 for stable contribution margin and returns, 5 for thin margin, high returns, fee changes or frequent discounting.
- Client decision cadence: 1 for monthly governance, 5 for weekly or daily trading decisions with multiple stakeholders.
A client scoring 8 can probably sit in a light-touch operating rhythm. A client scoring 18 needs senior review. A client scoring 23 should not be added to an account manager who is already carrying two other high-score clients, even if the calendar says there are “available hours”.
Example: Lakeview Home Goods in Ohio spends $32,000 per month across Amazon Sponsored Products, Walmart Connect and Google Shopping. The catalog has 620 SKUs, but only 40 drive most revenue. On a standard hours model, the account looks efficient: four hours of ads, two hours of reporting, one trading call. On the decision-load model, the score jumps because 18 of the top 40 SKUs have less than 21 days of stock and the Walmart return rate is 11% versus 6% on Amazon. The right capacity decision is not “add two reporting hours”. It is “give this client a weekly stock-and-margin review until inventory normalizes”.
The trade-off: utilization versus judgement
Agency owners love utilization because it looks objective. If everyone is at 82%, the business feels healthy. But marketplace work punishes teams that optimize for utilization alone. You can fill every hour and still miss the one decision that mattered.
There is a trade-off here. A marketplace agency needs enough process to scale, but enough slack to react to commercial events. Prime Day, Walmart deal events, bol promotions, Amazon fee changes, feed rejections and stock delays do not respect your perfect resource plan. If senior people are booked at 95% every week, urgent marketplace decisions get delegated to whoever has a spare half hour. That is how agencies create expensive mistakes while looking operationally disciplined.
My rule of thumb: protect 10-15% senior decision capacity for the clients with the highest commercial risk. Do not let that time become random client calls. Use it for profit reviews, escalation decisions and pre-emptive changes before the client sees the problem in next week’s report.
Where FiveX fits into the capacity model
FiveX helps agencies move capacity planning from “who has time?” to “which client needs a profit decision?”. That matters because marketplace data usually lives in too many places: Amazon Ads, Seller Central, Walmart, bol, Shopify, feed tools, inventory systems and finance exports. The team loses hours before the actual thinking starts.
First, FiveX creates a multi-client marketplace analytics view so account leads can compare sales, ad spend, SKU performance and operational signals across clients without rebuilding the spreadsheet every Monday. Portfolio visibility is the foundation for capacity planning because it shows where attention is commercially needed.
Second, FiveX connects advertising performance to profitability. ROAS alone is a dangerous workload trigger. A campaign at 4.0 ROAS may be fine for a 48% margin product and terrible for a 19% margin product after marketplace fees and returns. FiveX helps teams work with contribution margin, break-even ACOS, TACoS and product profitability so escalations are based on profit permission, not vanity efficiency.
Third, FiveX brings automation and AI recommendations into the operating rhythm. Agencies can use alerts and recommendations for budget pacing, margin guardrails, inventory pressure and product-level decisions. That does not replace the strategist. It gives the strategist a cleaner queue: these three clients need a decision; these seven are within guardrails; this one needs a client approval before spend increases.
Build the weekly agency operating rhythm
Software only improves capacity if it changes the rhythm of the agency. A practical weekly model looks like this:
- Monday portfolio triage: sort clients by capacity score, margin risk, stock risk and ad spend movement. Decide where senior attention goes before the calendar fills itself.
- Tuesday profit actions: adjust budgets, bids, campaign roles, feed priorities and marketplace focus for the clients with the highest commercial risk.
- Wednesday client decisions: turn internal findings into clear choices: replenish, pause, promote, reprice, shift channel budget or accept lower margin for a defined reason.
- Thursday automation review: check which alerts created good decisions and which ones caused noise. Tighten guardrails.
- Friday capacity review: compare planned workload with actual decision load. If a client repeatedly scores high, change scope, fee, cadence or staffing.
Example: Hudson Pet Supply pays a US marketplace agency $7,500 per month to manage Amazon, Walmart and TikTok Shop. The retainer looked profitable until TikTok creator spikes created three emergency stock decisions in two weeks. The agency had planned 22 monthly hours. Actual work became 37 hours, mostly because nobody had connected creator-driven demand to shared warehouse stock. With a capacity score, the agency can renegotiate the scope: TikTok Shop launch support becomes a separate operating package with inventory checks, creator approval rules and margin reporting. That protects the agency margin and gives the client a better service.
What to look for in marketplace agency capacity planning software
If you are evaluating software, do not only ask whether it has dashboards. Ask whether it reduces decision load. The checklist is straightforward:
- Can the platform separate clients cleanly while giving leadership one portfolio view?
- Can it connect marketplace sales, ad spend, inventory and product profitability?
- Can it show SKU-level margin, not just revenue and ROAS?
- Can alerts be based on commercial guardrails such as break-even ACOS, stock cover, Buy Box loss or return-rate movement?
- Can reporting be reused across clients without making every client look identical?
- Can account managers see which actions are urgent, which are optional and which require client approval?
- Can the agency use the data to adjust scope and pricing when a client’s decision load changes?
The last question is the one agencies skip. Capacity planning is not just a staffing exercise. It is a pricing exercise. If a client’s marketplace complexity doubles, the retainer should not stay the same because the dashboard is more automated. Automation removes repetitive work; it does not remove commercial responsibility.
The better agency promise
The best marketplace agencies will not win by saying, “We manage more channels.” Many agencies can say that now. They will win by saying, “We know where your profit is at risk, we know which decisions matter this week, and we have the operating system to act before the report is old news.”
That is a stronger promise for the client and a healthier model for the agency. It protects senior attention. It makes scope conversations factual. It turns software from a reporting cost into a capacity multiplier.
So before hiring another account manager, ask a sharper question: are we short on people, or are we using good people to find the work instead of make the decision? In marketplace agency work, that difference is often the margin.